A method for measuring the minimum clearance of a crossing line by using binocular vision technology

Abstract

The application discloses a kind of methods for measuring the minimum gap of intersection and crossing line using binocular vision technology, comprising the following steps: S1, selecting the structure parameters of suitable binocular vision system, calibrating binocular system;S2, hanging different markers on multiple suitable positions on intersection and crossing line, and collecting marker image using binocular camera;S3, using template matching algorithm, finding the pixel coordinates of marker point in marker image;S4, calculating the world coordinates of marker point, fitting the parameter equation of intersection and crossing line, and solving the parameter equation to obtain the minimum gap between intersection and crossing line.The application uses binocular measurement technology, accelerates the calculation process of template matching, fits intersection and crossing line with parameter equation, constructs optimization equation, realizes the rapid and intelligent measurement of the minimum gap of intersection and crossing line, and improves work efficiency.

Description

Technical Field

[0001] This invention belongs to the field of power transmission line measurement and monitoring technology, specifically relating to a method for measuring the minimum gap between crossing lines using binocular vision technology. Background Technology

[0002] With the development of power transmission projects in my country, the number of overhead transmission lines is increasing, and the phenomenon of crossing and spanning lines is also becoming more frequent. The minimum clearance between two lines is an issue that needs to be considered when crossing lines. If the clearance is too small, flashover may occur between the two lines, causing serious accidents and greatly threatening the safe and stable operation of overhead transmission lines.

[0003] Currently, the methods for measuring the minimum clearance at power line crossings mainly include manual visual inspection, rope measurement, theodolite, and total station, and there is a lack of unified and objective operating standards among these different methods. Furthermore, manual measurement methods not only rely on the condition and experience of the inspection personnel, but also have relatively low efficiency and accuracy. With the increasing scale of the power grid, manual measurement methods can no longer meet the coverage and immediacy requirements of power line inspections, necessitating an effective intelligent method to replace manual measurement.

[0004] A search revealed that patent CN 112964185 discloses an online monitoring system for transmission line crossing points based on binocular stereo vision. This patent discloses the system composition for monitoring transmission lines using binocular stereo vision and can effectively measure the spacing between crossing lines, but it does not disclose the process for achieving the minimum clearance measurement. Therefore, this invention aims to propose a method for measuring the minimum clearance between crossing lines for daily monitoring and maintenance of overhead transmission line crossing areas. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for measuring the minimum gap of crossing lines using binocular vision technology. This method uses binocular measurement technology to accelerate the calculation process of template matching, fit the crossing lines with parametric equations, and construct an optimized equation to achieve rapid and intelligent measurement of the minimum gap of crossing lines, thereby improving work efficiency.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for measuring the minimum gap between crossing lines using binocular vision technology, characterized in that: based on the known position of the suspension point of the crossing line, the crossing line is photographed using a binocular camera, and the minimum gap between the crossing lines is calculated, the steps of which are as follows:

[0007] S1. Select appropriate structural parameters for the binocular vision system and calibrate the binocular system.

[0008] S2, different markers are suspended at multiple suitable locations on the crossing line, and images of the markers are captured using a binocular camera;

[0009] S3 uses a template matching algorithm to find the pixel coordinates of the marker points in the marker image;

[0010] S4. Calculate the world coordinates of the marker point, fit the parametric equation of the crossing line, and solve the parametric equation to obtain the minimum gap between the crossing lines.

[0011] To further explain the above method, the suspension points of the marker include two suspension points on the two adjacent towers of the crossing line, and two suspension points on the two adjacent towers of the crossed line. The coordinates of these four suspension points in the world coordinate system are known.

[0012] To further explain the above method, the selection of the appropriate position is related to the position and angle of the image acquired by the binocular camera. The image acquired by the binocular camera contains all markers and is clearly visible.

[0013] To further explain the above method, the template matching algorithm uses the marker image as a template and performs template matching throughout the acquired image. The specific calculation equation for the response matrix R during the matching process is as follows:

[0014]

[0015] Where S is the acquired image matrix, T is the template image matrix, (N,M) is the size of the template image, and S ij This is a region at (i,j) in the acquired image that is the same size as the template image. This region can be represented by the equation:

[0016]

[0017] The solution to R(i,j) is based on the following convolution theorem:

[0018]

[0019] Where F() is the Fourier transform, F -1 () represents the inverse Fourier transform. Combining equations 1 and 2, we get:

[0020]

[0021] Where T' is the extended marker point template image matrix obtained by processing the marker point template image matrix T; S 2 It is the square of the elements of the acquired image S; I is the auxiliary calculation matrix; the position of the largest response matrix R is the pixel coordinate of the marker point in the adjacent traverse image.

[0022] To further explain the above method, the extended marker template image matrix T' is a zero-padding process performed on the n×m conductor marker template image T, so that the size of the processed extended marker template image T' is consistent with the size N×M of the adjacent conductor image S. The mathematical model is expressed as:

[0023]

[0024] To further explain the above method, the auxiliary calculation matrix I is constructed by building a matrix of size N×M, where the elements within the upper left corner (n×m) are 1s and the remaining elements are 0s. The mathematical model is expressed as follows:

[0025]

[0026] To further explain the above method, the world coordinates (x, y) of the marker point... w ,y w ,z w ) is based on the pixel coordinates (u) of the marker point on the left and right cameras. l ,v l ), (u r ,v r ) and the intrinsic and extrinsic parameters A of the stereo camera l A r [R] l t l ]、[R r t r The calculation equation is as follows:

[0027]

[0028] To further explain the above method, the process of constructing the parametric equations of the crossing line is as follows:

[0029] (1) Select a suitable, continuous, and smooth spatial curve equation that can approximately represent the attitude of the overhead conductor in space, such as a catenary or a parabolic curve. The curve equation can be expressed as:

[0030]

[0031] (2) Determine the spatial coordinates (x0, y0, z0) of the suspension points of the line on the two adjacent towers. n ,y n ,z n Substitute into the curve equation:

[0032] y0 = f1(x0)

[0033] z0 = f2(x0)

[0034] y n=f1(x n )

[0035] z n =f2(x n )

[0036] As constraints, some coefficients of the equations are determined or simplified, and the domain of the function is also determined.

[0037] To further explain the above method, the minimum gap between crossing lines is obtained by solving the parametric equations, based on the fitted parametric equations of the crossing line and the line being crossed:

[0038]

[0039]

[0040] The shortest distance Δ between the two intersecting lines can be expressed as:

[0041]

[0042] Where (X1,Y1,Z1) and (X2,Y2,Z2) are points on the two lines respectively; when X1 and X2 take values ​​within the domain that minimize the distance between the two points, this distance is the minimum distance between the two wires.

[0043] The beneficial effects of adopting the above technical solution are as follows: This invention utilizes binocular vision technology to measure the minimum gap between crossing lines. It introduces binocular vision technology and uses relevant theorems to accelerate the calculation process of finding the pixel coordinates of marker points on the image of the crossing line through template matching. By calculating in the frequency domain, the pixel coordinates of the marker points in the acquired image are quickly obtained, thus speeding up the measurement. Based on the spatial coordinates of the marker points of the crossing lines, the parametric equations of the crossing lines are fitted, and an optimized equation for the minimum distance between the crossing lines is constructed through the parametric equations. This enables accurate calculation of the minimum distance between the crossing lines and its corresponding position on the conductor, providing a new method for measuring the minimum gap between crossing lines. It is also used to determine whether the transmission line is in a safe operating state and to issue an early warning when the safe value is exceeded. Attached Figure Description

[0044] Figure 1 This is a flowchart of the algorithm of this invention;

[0045] Figure 2 This is a schematic diagram of the conductor structure to be measured;

[0046] Figure 3 This is a schematic diagram of a binocular vision system.

[0047] Figure 4 It is a camera imaging model. Detailed Implementation

[0048] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are merely descriptive and not limiting, and should not be used to limit the scope of protection of the present invention.

[0049] According to the appendix Figure 1 and 2 This invention discloses a method for measuring the minimum clearance between crossing lines using binocular vision technology. Based on the known location of the suspension points of the crossing lines, a binocular camera is used to photograph the crossing lines, and the minimum clearance is calculated. The method includes the following steps:

[0050] (1) Select appropriate structural parameters for the binocular vision system and calibrate the binocular system.

[0051] The structure of a binocular vision system is as follows Figure 3 As shown, the single camera is first calibrated to obtain the intrinsic parameter matrix A of the left and right cameras. l A r The extrinsic parameter matrices of the left and right cameras relative to the same coordinate system [R] l t l ]、[R r t r ], and the distortion coefficient matrix of the left and right cameras.

[0052] Using a certain camera coordinate system as the world coordinate system, taking the left camera coordinate system as an example, we can obtain the extrinsic parameter matrix [Rt] of the right camera relative to the left camera.

[0053] P cr =RP cl +t

[0054] in,

[0055]

[0056] t = t r -Rt l

[0057] According to the camera imaging model, there is a relationship between the pixel coordinates and physical coordinates of a point in space, as shown in the following formula.

[0058]

[0059] Based on the calibration results of the binocular system (see Table 1) and the pixel coordinates (u) of a point in space in the left and right images... l ,v l ), (u r ,v rIt can correct lens distortion and use the corrected pixel coordinates to solve for the three-dimensional physical coordinates of this point.

[0060] Table 1. Calibration results of the binocular system

[0061]

[0062] For any calibration point P on a set of marked images, its coordinates in the world coordinate system are P0. w and coordinates P in the left and right camera coordinate systems cl P cr The following relationship exists.

[0063]

[0064] For multiple sets of relative extrinsic parameters calculated from multiple sets of calibration images, the relative extrinsic parameters are optimized and solved using the least squares method.

[0065] (2) Hang different markers at multiple suitable locations along the crossing line. The markers are suspended at two points on the two adjacent towers of the crossing line and two points on the two adjacent towers of the crossed line. The coordinates of these four points in the world coordinate system are known. Then, images of the markers are acquired using a binocular camera. The selection of suitable locations is related to the position and angle of the images acquired by the binocular camera. The images acquired by the binocular camera contain all the markers and are clearly visible.

[0066] The acquired marker images should be symmetrically or as symmetrically as possible. The focal points of the two optical axes of the two cameras should be as close as possible to the target point, so that the target is imaged as centrally as possible. Experiments have verified that the angles α1 and α2 between the optical axes of the two cameras and the baseline are optimally between 35° and 45°, and the field of view angles ω1 and ω2... Ideally, the angle should be within ±25°.

[0067] (3) Use template matching algorithm to find the pixel coordinates of the marker points in the marker image.

[0068] Image processing is used to extract the images of the intersecting line captured by the left and right cameras, and the least squares method is used to fit the intersecting line to obtain the pixel coordinates of the center line of the intersecting line.

[0069] Take a point on the center line of the left image fitting, with pixel homogeneous coordinates m. l According to the epipolar constraint, the epipolar line at this point on the right image is...

[0070] l r =Fm l

[0071] Where F is the basic matrix, and in this system, the left camera coordinate system is set as the world coordinate system, i.e.

[0072]

[0073] t l =[0 0 0] T

[0074] R = R r

[0075] t = t r

[0076] Then the fundamental matrix F is

[0077]

[0078] Where, [t] × For vector t = [t x t y t z ] T The antisymmetric matrix, i.e.

[0079]

[0080] The point on the right image where the epipolar line corresponding to a point on the fitting center line of the left image intersects with the fitting center line of the right image is the corresponding point on the right image.

[0081] In this invention, the template matching algorithm uses the marker image as a template to perform template matching in the entire acquired image. The specific calculation process of the response matrix R in the matching process is as follows:

[0082] The traverse marker template image matrix T is processed by zero-padding the n×m traverse marker template image T, so that the size of the processed extended marker template image T' is consistent with the size N×M of the adjacent traverse image S. The mathematical model is expressed as:

[0083]

[0084] Construct an auxiliary computation matrix I, which is an N×M matrix. The elements within the top-left n×m range are 1s, and all other elements are 0s. The mathematical model is represented as follows:

[0085]

[0086] The response matrix R of the traverse marker template image matrix T on the adjacent traverse image matrix S is calculated using the following formula:

[0087]

[0088] In the formula, F() and F -1 () represent the two-dimensional Fourier transform and inverse Fourier transform of the matrix, respectively, which can be calculated using the Fast Fourier (Inverse) Transform. T' is the extended marker template image matrix obtained by processing the traverse marker template image matrix T. I is the auxiliary calculation matrix. The Hadamard product represents the element-wise multiplication of corresponding positions in matrices A and B. This represents the element of matrix A divided by the element at the corresponding position in matrix B.

[0089] (4) Calculate the world coordinates of the marker points, fit the parametric equations of the crossing lines, and solve the parametric equations to obtain the minimum gap between the crossing lines.

[0090] First, the three-dimensional world coordinates of the marker points in space are calculated using camera parameters. Then, based on the three-dimensional world coordinates of the suspension points at both ends of the traverse, a suitable parametric equation form for the traverse in space is constructed. The marker points are then substituted into this parametric equation to fit the undetermined coefficients.

[0091] Choosing a suitable, continuous, and smooth spatial curve equation that can approximately represent the attitude of the overhead conductor in space, such as a catenary or parabola, the curve equation can be expressed as:

[0092]

[0093] The spatial coordinates of the suspension points at both ends of the conductor are in Figure 2 Let M1(x0,y0,z0) and N1(X) be the two values ​​of the given information. n ,Y n Z n ) or (M2(x0,y0,z0) and N2(X n ,Y n Z n Substitute it into the curve equation:

[0094] y0 = f1(x0)

[0095] z0 = f2(x0)

[0096] y n =f1(x n )

[0097] z n =f2(x n )

[0098] As constraints, some coefficients of the equations are determined or simplified, and the domain of the function is also determined.

[0099] Using the parametric equations of two adjacent traverses, a constrained optimization equation for minimizing the distance between them is constructed. For two points (X1, Y1, Z1) and (X2, Y2, Z2) located on the two traverses respectively, the distance Δ between the two points can be expressed as:

[0100]

[0101] When X1 and X2 take values ​​within the domain that minimize the distance between the two points, that distance is the minimum distance between the two conductors.

[0102] By solving the optimization equation, we can obtain the two points corresponding to the minimum gap between the two conductors and the minimum gap value.

[0103] The response matrix R of the traverse marker template image matrix T on the adjacent traverse image matrix S is linearly related to the NCC template matching calculation result. The response positions of the points with the maximum response are the same. The derivation process is as follows:

[0104] The specific calculation method for NCC template matching is as follows:

[0105]

[0106] In the formula Removing it does not affect the determination of the position of the maximum response in the response matrix R, and it can simplify the formula, reduce the amount of calculation, and improve the speed.

[0107] Construct the processed extended marker template image matrix T' and auxiliary calculation matrix I:

[0108]

[0109]

[0110] The formula for calculating the response can be further transformed into:

[0111]

[0112] in, This indicates signal correlation operations.

[0113] Based on the similarity between convolution and correlation operations, and the convolution theorem, the formula for calculating the response in the frequency domain using Fast Fourier Transform is as follows:

[0114]

[0115] As can be seen from the above formula, the calculation result of this formula reduces the amount of computation by using the Fast Fourier Transform to change the complex spatial domain correlation operation into a simple frequency domain multiplication operation without reducing the accuracy.

[0116] Table 2 Comparison of the efficiency of the calculation methods in this invention

[0117]

[0118] As shown in Table 2, experimental comparisons reveal that this invention, through the convolution theorem, reduces computational complexity and significantly improves computational efficiency while maintaining accuracy. Furthermore, when calculating the response of a 25×35 conductor marker template image matrix T to an adjacent 720×1280 conductor image S, the computational speed of the proposed formula is far slower than previous methods, yet the results are almost identical.

[0119] Although embodiments and drawings of the present invention have been disclosed for illustrative purposes, those skilled in the art will understand that various substitutions, variations and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the scope of the present invention is not limited to the contents disclosed in the embodiments and drawings.

Claims

1. A method for measuring the minimum clearance between crossing lines using binocular vision technology, characterized in that: Based on the known locations of the suspension points of the crossing lines, the minimum clearance between the crossing lines is calculated by taking pictures of the crossing lines with a binocular camera. The steps are as follows: S1. Select appropriate structural parameters for the binocular vision system and calibrate the binocular system. S2, different markers are suspended at multiple suitable locations on the crossing line, and images of the markers are captured using a binocular camera; S3 uses a template matching algorithm to find the pixel coordinates of the marker points in the marker image; S4. Calculate the world coordinates of the marker point, fit the parametric equation of the crossing line, and solve the parametric equation to obtain the minimum gap between the crossing lines.

2. The method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 1, characterized in that: The suspension points of the marker include two suspension points on the two adjacent towers of the crossing line, and two suspension points on the two adjacent towers of the crossed line. The coordinates of these four suspension points in the world coordinate system are known.

3. The method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 1, characterized in that: The selection of a suitable location is related to the position and angle of the image captured by the binocular camera. The image captured by the binocular camera contains all markers and is clearly visible.

4. The method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 1, characterized in that: The template matching algorithm described above uses the marker image as a template and performs template matching throughout the entire acquired image. The specific calculation equation for the response matrix R during the matching process is as follows: （1） in, It is the acquired image matrix. It is a template image matrix. It is the size of the template image. In the acquired image A region defined by the size of the template image, which can be represented by the equation: The The solution is based on the following convolution theorem: （2） in, It is a Fourier transform. It is the inverse Fourier transform. Combining Equation 1 and Equation 2, we get: （3） in, For the marker point template image matrix The resulting extended marker template image matrix; It is to acquire images The square of the element; To assist in the calculation of the matrix; response matrix The largest position is the pixel coordinate of the marker point in the adjacent traverse image.

5. A method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 4, characterized in that: The extended marker point template image matrix It is for a size of Wire marker template image Zero-padding is performed to make the processed extended marker template image Size and adjacent conductor images Size Consistent, the mathematical model is expressed as: 。 6. A method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 4, characterized in that: The auxiliary calculation matrix The construction method is to construct a size of The matrix, top left corner Elements within the specified range are 1, and all other elements are 0. The mathematical model is represented as follows: 。 7. A method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 1, characterized in that: The world coordinates of the marker point It is based on the pixel coordinates of the marker points on the left and right cameras. , Internal and external parameters of binocular cameras , , , The calculation equation is as follows: 。 8. A method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 4, characterized in that: The process of constructing the parametric equations of the crossing line: (1) Select a suitable continuous and smooth spatial curve equation, and the curve equation can approximately represent the attitude of the overhead conductor in space. The curve equation can be expressed as: (2) Determine the spatial coordinates of the suspension points of the line on the two adjacent towers. Substitute into the curve equation: ； As constraints, some coefficients of the equations are determined or simplified, and the domain of the function is also determined.

9. A method for measuring the minimum gap between crossing lines using binocular vision technology according to claim 1, characterized in that: The minimum gap between crossing lines is obtained by solving the parametric equations, based on the fitted parametric equations of the crossing line and the line being crossed: Then the shortest distance between the two crossing lines. It can be represented as ;in, and These are points on the two lines respectively; when , The distance between two points is the minimum distance when the value is taken within the defined domain.

Images